1. Field of the Invention
The present invention relates to a recovery processing method for a device specific information which operates by developing device specific information, stored on a medium, onto a memory, and more particularly to a recovery processing method for device specific information for executing recovery processing for read errors of updated device specific information which are stored on a medium, and the medium storage device.
2. Description of the Related Art
Because of the recent demand for the computerized processing of data, a capacity increase is demanded for the medium storage devices of magnetic medium devices and optical medium devices for storing data. Therefore the track density and the recording density of medium storage devices are constantly increasing. In such medium storage devices, it is indispensable to monitor the state of the medium and the state of the read and write, so device specific information is stored in a medium which is a non-volatile memory for each device.
The device specific information includes information to be updated. In the case of a magnetic disk device, for example, if automatic replacement processing is generated, the addresses of an error sector (replacement source sector) and the replacement destination sector which took over the error sector are registered in the defective sector management information, and must be held even if power is turned OFF (see Japanese Patent Application Laid-Open No. H4-245072).
The SMART (Self Monitoring Analysis Report Technology) drive attribute values, which are guarantee information, includes information which is accumulated when the magnetic medium device is operating, such as the power ON time and the read/write error rate, and this information also must be held even if power is turned OFF.
In this way the magnetic disk device has information which is updated after shipment, and must be held even if power is turned OFF, and this information is stored as device specific information in a non-volatile memory or in the system area on the medium to which the user cannot access.
And when power is turned ON, information which has been updated/accumulated up to the point when power was turned OFF the last time is loaded, and this information is updated/accumulated. Therefore when power is turned ON, information must be read from the non-volatile memory or from the system area on the medium.
In such a magnetic disk device, an enormous size of data is required for the defective sector management information and the SMART information as the capacity of the medium increases. This data is indispensable for the normal execution of medium access and command operation of the magnetic disk device, and is stored in the device specific area on the medium.
The above data is read by the power ON sequence, resides in a memory, is updated during the operation of the magnetic disk device, and is stored in the device specific area of the medium regularly or at a predetermined timing, such as when a command is received.
Because of the recent popularization of mobile equipment, such a medium storage device is also mounted on mobile equipment (e.g. notebook personal computers, portable AV (Audio/Visual) equipment)). In this operating environment, write errors tend to occur in the medium storage device because of temperature changes and vibration. Therefore if device specific information is written and stored in the system area on the medium under such an environment, the device specific information may not be read when power is turned ON the next time.
For example, when vibration is applied to a disk device, an off-track motion may occur and the drive specific information may not be able to be accurately written to the target track. In this case, in the subsequent power ON sequence, the device specific information cannot be read even if the head seeks the target track.
In the same way, because of recording density which is becoming increasingly dense lately, an off-track motion tends to occur when the temperature is different, and, for example, if the temperature changes between the write and read of the device specific information, a write error occurs when a device specific information is stored, and if power is shut OFF in this status, an error occurs in reading the device specific information in the subsequent power ON sequence.
The device specific information of the disk device includes information which is indispensable for executing commands, and if this information cannot be read during the power ON sequence, the disk device may not operate normally. Also if the device specific information which was read with error is defective management information, normal medium access cannot be guaranteed.
In other words, if the defective sector management information cannot be read in a device where automatic replacement processing is generating, not the replacement destination sector but the replacement source sector is accessed. The replacement source sector is the sector where an error originally occurred, and the data has been written to the replacement destination sector, so a read error or a data comparison error occurs. Also if the defective sector information registered at shipment is included in the defective sector management information, accessing the defective sector at shipment may occur.
In the same way, if the SMART drive attribute information cannot be read, commands cannot be executed since the attribute values to be reported to the host by the SMART commands do not exist.
Once the disk device enters the above mentioned status, normal operation cannot be guaranteed, so a self diagnosis error or an abort of the received command occurs. Also if the disk device ended all the commands issued from the host system as an abort or error, the host system may not be started up, then the backup operation for the data stored in the disk device cannot be executed either.